Bug#850284: reprotest: add a mode to automatically attempt to detect which variations cause unreproducibility

[Holger Levsen]

On Thu, Jan 05, 2017 at 06:07:58PM +0100, Ximin Luo wrote:
> One idea that was brought up at RWS2 in Berlin (I forgot who came up with it
> now, sorry) is to add a mode to automatically attempt to detect which
> variations cause a package to be unreproducible.
 
kudos for that idea, again! :)

> Note that this is not as simple as a "bisect" algorithm which assumes that the
> items to be tested are in some linear order. In fact, it's unclear if we can
> make the algorithm (using the above example) more efficient than the brute
> force method of "test a, test b, test c".

I'm not fully sure how to take this into consideration, but some
variations are more likely to cause unreproducible builds than others,
IOW the ordering of the test shouldnt be random, but rather follow such
(assumed) probability.


-- 
cheers,
Holger


signature.asc
Description: Digital signature
___
Reproducible-builds mailing list
Reproducible-builds@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds

Bug#850284: reprotest: add a mode to automatically attempt to detect which variations cause unreproducibility

[Ximin Luo]

Package: reprotest
Version: 0.4
Severity: wishlist

Dear Maintainer,

One idea that was brought up at RWS2 in Berlin (I forgot who came up with it
now, sorry) is to add a mode to automatically attempt to detect which
variations cause a package to be unreproducible.

For example, given variations {a, b, c}, if varying none of these results in a
reproducible build, but varying all of them makes it unreproducible, then
reprotest could automatically try to figure out which combinations of {a, b, c}
"cause" the unreproducibility.

Note that this is not as simple as a "bisect" algorithm which assumes that the
items to be tested are in some linear order. In fact, it's unclear if we can
make the algorithm (using the above example) more efficient than the brute
force method of "test a, test b, test c".

X

-- System Information:
Debian Release: stretch/sid
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'unstable-debug'), (500, 
'testing-debug'), (300, 'unstable'), (200, 'experimental'), (1, 
'experimental-debug')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.8.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_GB.utf8, LC_CTYPE=en_GB.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages reprotest depends on:
ii  apt-utils  1.4~beta2
ii  diffoscope 63
ii  libdpkg-perl   1.18.18
ii  procps 2:3.3.12-3
ii  python3-debian 0.1.29
ii  python3-pkg-resources  32.0.0-1
pn  python3:any

Versions of packages reprotest recommends:
ii  disorderfs   0.5.1-1
ii  locales-all  2.24-8

Versions of packages reprotest suggests:
ii  autodep8 0.8
ii  qemu-system  1:2.7+dfsg-3+b1
ii  qemu-utils   1:2.7+dfsg-3+b1
ii  schroot  1.6.10-2+b2

-- no debconf information

___
Reproducible-builds mailing list
Reproducible-builds@lists.alioth.debian.org
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/reproducible-builds